System and method of analyzing health care provider networks

Abstract

A computer system and method for creating a network model of health care providers comprising individuals that are connected to a particular influencer or health care provider. The system determines an influence rank that is assigned to each network that corresponds to the level of influence in comparison to other networks in a cohort. For example, the influence rank could be based, at least in part, on the number of patients treated by a health care provider and that provider's patient referrals. In some cases, the system provides a graphical user interface through which the networks can be visualized. These networks can be filtered based on a variety of categories to pinpoint specific networks, such as by geographic regions, specialties, diseases, and/or specific health care providers.

What is claimed is: 
     
         1 . An apparatus comprising:
 a storage device; and   at least one processor coupled to the storage device, wherein the storage device stores a program for controlling the at least one processor, and wherein the at least one processor, being operative with the program, is configured to:
 obtain historical claims data representative of encounters between a plurality of patients and health care providers (“HCPs”), including referrals between HCPs; 
 analyze the historical claims data to create a network model representative of connections between HCPs; 
 determine an influence score for the HCPs that represents a level of influence of a respective HCP in comparison to other HCPs that are connected to that respective HCP; and 
 transmit data representative of at least a portion of the network model with the influence score of those HCPs in that portion of the network. 
   
     
     
         2 . The apparatus of  claim 1 , wherein at least a portion of the connections in the network model are based on referrals between HCPs. 
     
     
         3 . The apparatus of  claim 1 , wherein the influence score is determined based at least in part on a number of patients treated directly by a HCP. 
     
     
         4 . The apparatus of  claim 1 , wherein the influence score is determined based at least in part on patients of a selected HCP treated by other HCPs who are directly influenced by the selected HCP. 
     
     
         5 . The apparatus of  claim 4 , wherein the influence score is determined based at least in part on patients of a selected HCP treated by other HCPs who are directly or indirectly influenced by the selected HCP. 
     
     
         6 . The apparatus of  claim 1 , wherein the influence score is determined based at least in part on a sum of patients treated and patients indirectly influenced by a selected HCP. 
     
     
         7 . The apparatus of  claim 1 , wherein the influence score is determined based at least in part on a number of HCPs influenced by a selected HCP either directly or indirectly regardless of how many patients treated by the selected HCP. 
     
     
         8 . The apparatus of  claim 1 , wherein the program is configured to cause the processor to generate an interface from which the network model representative of connections between patients and HCPs can be viewed. 
     
     
         9 . The apparatus of  claim 8 , wherein the interface includes a data table representative of one or more criteria that comprise the network model. 
     
     
         10 . The apparatus of  claim 8 , wherein the interface includes a graphical visualization of the network model. 
     
     
         11 . The apparatus of  claim 10 , wherein the graphical visualization includes one or more nodes representative of respective HCP and with connections between the nodes representing connected HCPs in the network model. 
     
     
         12 . The apparatus of  claim 10 , wherein the graphical visualization is configured to visually differentiate nodes based on relative influence. 
     
     
         13 . The apparatus of  claim 12 , wherein the visual differentiation between nodes based on relative influence is one or more of relative size and/or color of the nodes to identify relative influence between nodes. 
     
     
         14 . The apparatus of  claim 11 , wherein responsive to detection of an input device hovering over a hovered node, the interface displays a name of the HCP corresponding to the hovered node. 
     
     
         15 . The apparatus of  claim 14 , wherein responsive to selection of a selected node, the interface generates a network of the HCP corresponding to the selected node. 
     
     
         16 . The apparatus of  claim 11 , wherein responsive to detection of an input device hovering over a hovered connection between nodes, the interface generates an influence score between the HCPs corresponding to the hovered connection. 
     
     
         17 . The apparatus of  claim 8 , wherein the interface includes one or more filters configured to remove information of the network model from the interface based on user-selected filter criteria. 
     
     
         18 . The apparatus of  claim 17 , wherein the one or more filters are configured to allow selection of one or more of the following as filter criteria: geographic regions, specialties, diseases, and/or specific health care providers. 
     
     
         19 . A computer-implemented method, comprising:
 obtaining historical claims data representative of encounters between a plurality of patients and health care providers (“HCPs”), including referrals between HCPs;   analyzing the historical claims data to create a network model representative of connections between HCPs;   determining an influence score for the HCPs that represents a level of influence of a respective HCP in comparison to other HCPs that are connected to that respective HCP; and   transmitting data representative of at least a portion of the network model with the influence score of those HCPs in that portion of the network.   
     
     
         20 . The method of  claim 19 , wherein at least a portion of the connections in the network model are based on referrals between HCPs. 
     
     
         21 . The method of  claim 19 , wherein the influence score is determined based at least in part on a number of patients treated directly by a HCP. 
     
     
         22 . The method of  claim 19 , wherein the influence score is determined based at least in part on patients of a selected HCP treated by other HCPs who are directly influenced by the selected HCP. 
     
     
         23 . The method of  claim 22 , wherein the influence score is determined based at least in part on patients of a selected HCP treated by other HCPs who are directly or indirectly influenced by the selected HCP. 
     
     
         24 . The method of  claim 19 , wherein the influence score is determined based at least in part on a sum of patients treated and patients indirectly influenced by a selected HCP. 
     
     
         25 . The method of  claim 19 , wherein the influence score is determined based at least in part on a number of HCPs influenced by a selected HCP either directly or indirectly regardless of how many patients treated by the selected HCP. 
     
     
         26 . The method of  claim 19 , wherein the program is configured to cause the processor to generate an interface from which the network model representative of connections between patients and HCPs can be viewed. 
     
     
         27 . The method of  claim 26 , wherein the interface includes a data table representative of one or more criteria that comprise the network model. 
     
     
         28 . The method of  claim 26 , wherein the interface includes a graphical visualization of the network model. 
     
     
         29 . The method of  claim 28 , wherein the graphical visualization includes one or more nodes representative of respective HCP and with connections between the nodes representing connected HCPs in the network model. 
     
     
         30 . The method of  claim 28 , wherein the graphical visualization is configured to visually differentiate nodes based on relative influence. 
     
     
         31 . The method of  claim 30 , wherein the visual differentiation between nodes based on relative influence is one or more of relative size and/or color of the nodes to identify relative influence between nodes. 
     
     
         32 . The method of  claim 29 , wherein responsive to detection of an input device hovering over a hovered node, the interface displays a name of the HCP corresponding to the hovered node. 
     
     
         33 . The method of  claim 30 , wherein responsive to selection of a selected node, the interface generates a network of the HCP corresponding to the selected node. 
     
     
         34 . The method of  claim 29 , wherein responsive to detection of an input device hovering over a hovered connection between nodes, the interface generates an influence score between the HCPs corresponding to the hovered connection. 
     
     
         35 . The method of  claim 29 , wherein the interface includes one or more filters configured to remove information of the network model from the interface based on user-selected filter criteria. 
     
     
         36 . The method of  claim 33 , wherein the one or more filters are configured to allow selection of one or more of the following as filter criteria: geographic regions, specialties, diseases, and/or specific health care providers. 
     
     
         37 . A tangible, non-transitory computer readable medium storing instructions that, when executed by at least one processor, causes the at least one processor to perform a method comprising:
 obtaining historical claims data representative of encounters between a plurality of patients and health care providers (“HCPs”), including referrals between HCPs;   analyzing the historical claims data to create a network model representative of connections between HCPs;   determining an influence score for the HCPs that represents a level of influence of a respective HCP in comparison to other HCPs that are connected to that respective HCP; and   transmitting data representative of at least a portion of the network model with the influence score of those HCPs in that portion of the network.